SPIRE

SPIRE integrates planetary-scale metagenomic datasets to enable analysis of microbial diversity, taxonomy, and function.

Key Features:

• Dataset scope: Aggregates 99,146 metagenomic samples from 739 studies covering diverse habitats, geographic locations, and phylogenetic relationships.
• Assembly scale: Contains a total metagenomic assembly size of 16 terabases (Tbp).
• Predicted proteins: Includes 35 billion predicted protein sequences derived from metagenomic assemblies.
• Metagenome-assembled genomes (MAGs): Reconstructs 1.16 million novel MAGs classified as medium or high quality.
• Species-level clustering: Forms 92,134 species-level clusters from MAGs, with a majority remaining unclassified at the species level using existing tools.
• Taxonomic profiling: Provides taxonomic profiling via an updated custom mOTUs database.
• Functional annotation: Supplies multiple layers of functional annotation for metagenome-derived sequences.
• Cross-references: Links metagenomic data to external biological databases and cross-references the proGenomes3 high-quality genome reference.

Scientific Applications:

• Microbial diversity mapping: Enables planetary-scale surveys of microbial community composition across habitats and regions.
• Taxonomic expansion: Supports discovery and characterization of novel species-level clusters beyond current classifications.
• Taxonomic profiling: Facilitates community profiling using the updated custom mOTUs database.
• Functional potential analysis: Allows interrogation of metabolic and functional potential via multiple annotation layers.
• Comparative genomics: Enables genomic comparisons and contextualization through cross-references to proGenomes3 and other databases.

Methodology:

Metagenomic assemblies totaling 16 Tbp from 99,146 samples were generated, 35 billion protein sequences were predicted, 1.16 million MAGs were reconstructed and classified as medium or high quality, MAGs were clustered into 92,134 species-level groups, taxonomic profiling was performed using an updated custom mOTUs database, multiple layers of functional annotation were applied, and data were cross-referenced to proGenomes3 and other biological databases.